TFT-LCD Spot-Type Defect Detection in Module Process
In this paper we present a global approach for the automatic inspection of spot-type defects in TFT-LCD module process. The proposed method is based on a global image reconstruction scheme using the Fourier transform. Since TFT-LCD surface has the statistical and periodic texture, the isotropic spread of frequency components in the power spectrum space forms the shape approximate to a circle. By computing the convolution between the Fourier transform result and a ideal low pass filter (ILPF) , we can remove the TFT-LCD periodic and repetitive patterns using the inverse Fourier transform. In the restored image, the global structurally textured background is eliminated and the spot-type defects are preserved. This converts the difficult defect detection in TFT-LCD images into a simple segmentation problem. The spot-type defects are classified into general point defects and spot Mura defects .The experimental results from point defects and spot Mura defects have shown the efficacy of the proposed method.